Pritish K. Bhattacharyya

Clinical decision support systems for improving diagnostic accuracy and achieving precision medicine.

As research laboratories and clinics collaborate to achieve precision medicine, both communities are required to understand mandated electronic health/medical record (EHR/EMR) initiatives that will be fully implemented in all clinics in the United States by 2015. Stakeholders will need to evaluate current record keeping practices and optimize and standardize methodologies to capture nearly all information in digital format. Collaborative efforts from academic and industry sectors are crucial to achieving higher efficacy in patient care while minimizing costs. Currently existing digitized data and information are present in multiple formats and are largely unstructured. In the absence of a universally accepted management system, departments and institutions continue to generate silos of information. As a result, invaluable and newly discovered knowledge is difficult to access. To accelerate biomedical research and reduce healthcare costs, clinical and bioinformatics systems must employ common data elements to create structured annotation forms enabling laboratories and clinics to capture sharable data in real time. Conversion of these datasets to knowable information should be a routine institutionalized process. New scientific knowledge and clinical discoveries can be shared via integrated knowledge environments defined by flexible data models and extensive use of standards, ontologies, vocabularies, and thesauri. In the clinical setting, aggregated knowledge must be displayed in user-friendly formats so that physicians, non-technical laboratory personnel, nurses, data/research coordinators, and end-users can enter data, access information, and understand the output. The effort to connect astronomical numbers of data points, including ‘-omics’-based molecular data, individual genome sequences, experimental data, patient clinical phenotypes, and follow-up data is a monumental task. Roadblocks to this vision of integration and interoperability include ethical, legal, and logistical concerns. Ensuring data security and protection of patient rights while simultaneously facilitating standardization is paramount to maintaining public support. The capabilities of supercomputing need to be applied strategically. A standardized, methodological implementation must be applied to developed artificial intelligence systems with the ability to integrate data and information into clinically relevant knowledge. Ultimately, the integration of bioinformatics and clinical data in a clinical decision support system promises precision medicine and cost effective and personalized patient care.

Germinal Center and Activated B-Cell Profiles Separate Burkitt Lymphoma and Diffuse Large B-Cell Lymphoma in AIDS and Non-AIDS Cases

Morphologic features of Burkitt lymphoma (BL) and diffuse large B-cell lymphoma (DLBCL) overlap. No single phenotypic marker or molecular abnormality is pathognomonic. We tested a panel of 8 germinal center (GC) and activated B-cell (ABC) markers for their ability to separate BL and DLBCL. We diagnosed 16 BL and 39 DLBCL cases from 21 patients with AIDS and 34 without AIDS based on traditional morphologic criteria, Ki-67 proliferative index, and c-myc rearrangement (fluorescence in situ hybridization). After immunohistochemically staining tissue microarrays of BL and DLBCL for markers of GC (bcl-6, CD10, cyclin H) and ABC (MUM1, CD138, PAK1, CD44, bcl-2), we scored each case for the percentage of positive cells. Hierarchical clustering yielded 2 major clusters significantly associated with morphologic diagnosis (P < .001). For comparison, we plotted the sum of the GC scores and ABC scores for each case as x and y data points. This revealed a high-GC/low-ABC group and a low-GC/high-ABC group that were associated significantly with morphologic diagnosis (P < .001). Protein expression of multiple GC and ABC markers can separate BL and DLBCL.

Addition of lenalidomide to rituximab, ifosfamide, carboplatin, etoposide (RICER) in first-relapse/primary refractory diffuse large B-cell lymphoma

Relapsed/refractory diffuse large B‐cell lymphoma (DLBCL) is associated with a poor prognosis. Outcomes are particularly poor following immunochemotherapy failure or relapse within 12 months of induction. We conducted a Phase I/II trial of lenalidomide plus RICE (rituximab, ifosfamide, carboplatin, and etoposide) (RICER) as a salvage regimen for first‐relapse or primary refractory DLBCL. Dose‐escalated lenalidomide was combined with RICE every 14 d. After three cycles of RICER, patients with chemosensitive disease underwent stem cell collection and consolidation with BEAM [BCNU (carmustine), etoposide, cytarabine, melphalan] followed by autologous stem cell transplantation (autoSCT). Patients who recovered from autoSCT toxicities within 90 d initiated maintenance treatment with lenalidomide 25 mg daily for 21 d every 28 d for 12 months. No dose‐limiting or unexpected toxicities occurred with lenalidomide 25 mg plus RICE. Grade 3/4 haematological toxicities resolved appropriately, and planned dose density and dose intensity of RICER were preserved. No lenalidomide or RICE dose reductions were required in any of the three cycles. After two cycles of RICER, nine of 15 patients (60%) achieved a complete response, and two achieved a partial response (13%). Combining lenalidomide with RICE is feasible, and results in promising response rates (particularly complete response rates) in high‐risk DLBCL patients.

Reduced‐dose fludarabine, cyclophosphamide, and rituximab (FCR‐Lite) plus lenalidomide, followed by lenalidomide consolidation/maintenance, in previously untreated chronic lymphocytic leukemia

Fludarabine, cyclophosphamide, and rituximab (FCR) remains the standard of care for fit chronic lymphocytic leukemia (CLL) patients requiring first therapy. However, side effects can be significant, and patients with poor risk features have inferior outcomes. The purpose of this study was to evaluate reduced‐dose FCR (FCR‐Lite) plus lenalidomide (FCR2) followed by lenalidomide maintenance as a strategy to shorten immunochemotherapy in untreated CLL. Patients received four to six cycles of FCR2. Patients who were minimal residual disease (MRD) negative in peripheral blood (PB) and bone marrow (BM) initiated 12 months of lenalidomide maintenance after either four or six cycles (based on MRD status). The primary study endpoint was the complete response (CR) rate after four cycles of FCR2. Twenty patients were evaluable. After four cycles of FCR2, response rates were: CR, 45.0%; CR with incomplete blood count recovery (CRi), 5.0%; partial response (PR), 45.0%; and stable disease (SD), 5.0%. BM and PB samples from 27.8% and 52.9% of patients, respectively, were MRD negative. After six cycles, response rates were: CR, 58.3%; CRi, 16.7%; and PR, 25.0%. BM and PB samples from 50.0% and 72.7% of patients, respectively, were MRD negative. Overall, 75% of evaluable patients achieved a CR or CRi following FCR2. After 17.4 months of median follow‐up, one progression and one death occurred. Our findings suggest that FCR2 combines encouraging clinical activity with acceptable toxicity in previously untreated CLL. Lenalidomide can be safely added to FCR and may reduce chemotherapy exposure without compromising outcomes. Am. J. Hematol. 90:487–492, 2015.

Fibroblast growth factor-2 (FGF2) and syndecan-1 (SDC1) are potential biomarkers for putative circulating CD15+/CD30+ cells in poor outcome Hodgkin lymphoma patients.

BackgroundHigh risk, unfavorable classical Hodgkin lymphoma (cHL) includes those patients with primary refractory or early relapse, and progressive disease. To improve the availability of biomarkers for this group of patients, we investigated both tumor biopsies and peripheral blood leukocytes (PBL) of untreated (chemo-naïve, CN) Nodular Sclerosis Classic Hodgkin Lymphoma (NS-cHL) patients for consistent biomarkers that can predict the outcome prior to frontline treatment.Methods and materialsBioinformatics data mining was used to generate 151 candidate biomarkers, which were screened against a library of 10 HL cell lines. Expression of FGF2 and SDC1 by CD30+ cells from HL patient samples representing good and poor outcomes were analyzed by qRT-PCR, immunohistochemical (IHC), and immunofluorescence analyses.ResultsTo identify predictive HL-specific biomarkers, potential marker genes selected using bioinformatics approaches were screened against HL cell lines and HL patient samples. Fibroblast Growth Factor-2 (FGF2) and Syndecan-1 (SDC1) were overexpressed in all HL cell lines, and the overexpression was HL-specific when compared to 116 non-Hodgkin lymphoma tissues. In the analysis of stratified NS-cHL patient samples, expression of FGF2 and SDC1 were 245 fold and 91 fold higher, respectively, in the poor outcome (PO) group than in the good outcome (GO) group. The PO group exhibited higher expression of the HL marker CD30, the macrophage marker CD68, and metastatic markers TGFβ1 and MMP9 compared to the GO group. This expression signature was confirmed by qualitative immunohistochemical and immunofluorescent data. A Kaplan-Meier analysis indicated that samples in which the CD30+ cells carried an FGF2+/SDC1+ immunophenotype showed shortened survival. Analysis of chemo-naive HL blood samples suggested that in the PO group a subset of CD30+ HL cells had entered the circulation. These cells significantly overexpressed FGF2 and SDC1 compared to the GO group. The PO group showed significant down-regulation of markers for monocytes, T-cells, and B-cells. These expression signatures were eliminated in heavily pretreated patients.ConclusionThe results suggest that small subsets of circulating CD30+/CD15+ cells expressing FGF2 and SDC1 represent biomarkers that identify NS-cHL patients who will experience a poor outcome (primary refractory and early relapsing).

Alpha- and beta-synucleins are new diagnostic tools for acute erythroid leukemia and acute megakaryoblastic leukemia.

α-Synuclein is a key component of the Lewy body, a large globular protein complex that forms in the nervous system of patients with Parkinson disease and other dementias [1-3]. Since α-synuclein also occurs in megakaryocytic and erythroid lineages [4-7], we wondered what role synucleins had in the hematopoietic system. Therefore, we studied the expression of α-, β-, and γ-synucleins in a comprehensive panel of patient bone marrows and leukemic cell lines. We observed under expression of α-synuclein in the megakaryocytes of myeloproliferative neoplasm (MPN), but not normal reactive marrow (NRM) or myelodysplastic syndrome (MDS). Conversely, we observed over expression of β-synuclein in the blasts of megakaryoblastic leukemias (MegL), but not acute myeloid leukemia (AML) or erythroleukemia (EryL), suggesting that α- and β-synucleins could be useful adjunct markers for the early detection of MDS and the differential diagnosis of EryL and MegL from other AMLs.

Spectrin isoforms: differential expression in normal hematopoiesis and alterations in neoplastic bone marrow disorders.

Spectrins are large, rod-like, multifunctional molecules that participate in maintaining cell structure, signal transmission, and DNA repair. Because little is known about the role of spectrins in normal hematopoiesis and leukemogenesis, we immunohistochemically stained bone marrow biopsy specimens from 81 patients for αI, αII, βI, and βII spectrin isoforms in normal reactive marrow (NRM), myelodysplastic syndrome, myeloproliferative neoplasm, acute myeloid leukemia (AML) with well-characterized cytogenetic abnormalities, acute erythroid leukemia (EryL), and acute megakaryoblastic leukemia (MegL). In NRM, spectrin isoforms were differentially expressed according to cell lineage: αI and βI in erythroid precursors; αII and βII in granulocytes; and βI and βII in megakaryocytes. In contrast, 18 (44%) of 41 AMLs lacked αII spectrin and/or aberrantly expressed βI spectrin (P = .0398; Fisher exact test) and 5 (100%) of 5 EryLs expressed βII spectrin but lacked βI spectrin. The frequent loss and/or gain of spectrin isoforms in AMLs suggests a possible role for spectrin in leukemogenesis.

A unique case of merkel cell carcinoma and chronic lymphocytic leukaemia presenting in a single cutaneous lesion (collision tumour)

A 58-year-old man with 10 year history of chronic lymphocytic leukaemia (CLL) presented with a bulging mass (3.2×1.6×1.5 cm) in the right anterior abdominal wall. On microscopic examination the mass was found to be an epithelial neoplasm in the background of lymphoid proliferation. The epithelial cells were of moderate size, with scant cytoplasm and round nuclei, forming glandular, alveolar or sheet-like structures. These cells were immunoreactive for cytokeratin 20, chromogranin and synaptophysin. The above findings supported a diagnosis of Merkel cell carcinoma (MCC). The background small lymphocytes expressed phenotypic markers of B cells (CD20) and CD5, consistent with his known diagnosis of CLL. The incidence of a patient with a known history of CLL who develops a secondary MCC is rare. This is believed to be the second case report of a single lesion containing CLL and MCC.

Rituximab, cyclophosphamide-fractionated, vincristine, doxorubicin and dexamethasone alternating with rituximab, methotrexate and cytarabine overcomes risk features associated with inferior outcomes in treatment of newly diagnosed, high-risk diffuse large B-cell lymphoma.

Abstract Subtypes of diffuse large B-cell lymphoma (DLBCL) that have inferior outcomes after front-line therapy with R-CHOP (rituximab, cyclophosphamide, doxorubicin, vincristine and prednisone) have been identified. While it is agreed that R-CHOP is probably not adequate in these patients, there is no standard treatment approach for patients with DLBCL with high-risk features. We present results of a retrospective cohort study of high-risk DLBCL (defined as having at least one unfavorable risk factor: non-germinal center [GC] subtype by immunohistochemistry [IHC], Ki-67 ≥ 80%, high International Prognostic Index [IPI], c-MYC rearrangement) treated with R-HCVAD/R-MTX-AraC (rituximab, cyclophosphamide-fractionated, vincristine, doxorubicin and dexamethasone alternating with rituximab, methotrexate and cytarabine; R-HCVAD) as front-line therapy. With a median follow-up of 25.3 months, the 3-year PFS and OS estimates are 79% (95% confidence interval [CI], 65–88%) and 76% (95% CI, 61–86%), respectively, which are higher than those for historical comparisons with R-CHOP data for high-risk patients. These data are in accord with other recent reports of dose-intense front-line therapy of high-risk DLBCL. This analysis represents the largest reported cohort of patients with DLBCL treated with R-HCVAD. These data suggest that R-HCVAD can overcome traditional poor risk features such as high IPI, high Ki-67 and non-GC IHC pattern. Future work will focus on identifying molecular markers for failure in patients with DLBCL treated with dose-intensive regimens.

STAG2 Is a Biomarker for Prediction of Recurrence and Progression in Papillary Non–Muscle-Invasive Bladder Cancer

Purpose: Most bladder cancers are early-stage tumors known as papillary non–muscle-invasive bladder cancer (NMIBC). After resection, up to 70% of NMIBCs recur locally, and up to 20% of these recurrences progress to muscle invasion. There is an unmet need for additional biomarkers for stratifying tumors based on their risk of recurrence and progression. We previously identified STAG2 as among the most commonly mutated genes in NMIBC and provided initial evidence in a pilot cohort that STAG2-mutant tumors recurred less frequently than STAG2 wild-type tumors. Here, we report a STAG2 biomarker validation study using two independent cohorts of clinically annotated papillary NMIBC tumors from the United States and Europe. Experimental Design: The value of STAG2 immunostaining for prediction of recurrence was initially evaluated in a cohort of 82 patients with papillary NMIBC (“Georgetown cohort”). Next, the value of STAG2 immunostaining for prediction of progression to muscle invasion was evaluated in a progressor-enriched cohort of 253 patients with papillary NMIBC (“Aarhus cohort”). Results: In the Georgetown cohort, 52% of NMIBC tumors with intact STAG2 expression recurred, whereas 25% of STAG2-deficient tumors recurred (P = 0.02). Multivariable analysis identified intact STAG2 expression as an independent predictor of recurrence (HR = 2.4; P = 0.05). In the progressor-enriched Aarhus cohort, 38% of tumors with intact STAG2 expression progressed within 5 years, versus 16% of STAG2-deficient tumors (P < 0.01). Multivariable analysis identified intact STAG2 expression as an independent predictor of progression (HR = 1.86; P = 0.05). Conclusions: STAG2 IHC is a simple, binary, new assay for risk stratification in papillary NMIBC. Clin Cancer Res; 24(17); 4145–53. ©2018 AACR.